A method of fatigue grading and calling multi-dimension sensory for fatigue intervention

Abstract

The application relates to a method for fatigue grading and calling multidimensional senses for fatigue intervention, belonging to the field of fatigue driving detection in vehicle engineering, and comprising the following steps: acquiring fatigue state data of a vehicle driver; determining a corresponding fatigue grade of the vehicle driver according to the fatigue state data of the vehicle driver; determining a corresponding sense stimulation intervention zone group according to the corresponding fatigue grade of the vehicle driver; generating a corresponding sense stimulation strategy according to the sense stimulation intervention zone group; controlling a corresponding sense stimulation execution module in the vehicle to execute the sense stimulation strategy; acquiring fatigue state data after sense stimulation intervention; re-determining the corresponding fatigue grade of the vehicle driver according to the re-acquired fatigue state data; determining a current sense stimulation intervention effect according to the change of the fatigue grade before intervention and the fatigue grade after intervention; and adjusting a subsequent sense stimulation intervention zone group based on the current sense stimulation intervention effect. The application can realize differential and dynamic intervention according to the fatigue degree of the vehicle driver.

Description

Technical Field

[0001] This invention belongs to the field of fatigue driving detection technology in vehicle engineering, specifically involving a method for fatigue grading and using multi-dimensional senses for fatigue intervention. Background Technology

[0002] With the development of vehicle intelligence and cockpit sensing technology, fatigue driving detection has been gradually applied to the field of driving safety. Existing fatigue driving detection solutions typically determine driver fatigue based on information such as eye status, facial features, vehicle driving status, or steering wheel operation, and then provide alerts through audible alarms, light prompts, or seat vibrations. However, existing fatigue driving detection solutions suffer from problems such as insufficiently precise fatigue level classification, limited intervention methods, inadequate matching of intervention intensity to fatigue severity, and a lack of post-intervention feedback and re-evaluation mechanisms. These issues make it difficult for the system to provide differentiated and dynamic interventions based on the driver's fatigue level.

[0003] Therefore, it is necessary to propose a scheme that can classify fatigue and use multiple senses such as vision, hearing, smell, touch, and taste to intervene in fatigue according to different fatigue levels. Summary of the Invention

[0004] To address the problems existing in the prior art, this invention provides a method for fatigue grading and using multi-dimensional senses for fatigue intervention.

[0005] The technical solution adopted by this invention to solve the technical problem is as follows:

[0006] The first objective of this invention is to provide a method for fatigue grading and using multi-dimensional senses for fatigue intervention.

[0007] This invention provides a method for fatigue grading and multi-dimensional sensory intervention, which mainly includes the following steps:

[0008] Step 1: Fatigue detection grading based on multi-source features and fuzzy comprehensive evaluation;

[0009] Acquire driver fatigue data; determine driver fatigue level based on driver fatigue data;

[0010] Step 2: Multi-sensory mobilization based on fatigue level;

[0011] The corresponding sensory stimulation intervention group is determined based on the fatigue level of the vehicle driver; the corresponding sensory stimulation strategy is generated based on the sensory stimulation intervention group; and the corresponding sensory stimulation execution module in the vehicle is controlled to execute the sensory stimulation strategy.

[0012] Step 3: Reassess based on feedback of decreased fatigue level;

[0013] Acquire fatigue status data after sensory stimulation intervention; redetermine the fatigue level of the vehicle driver based on the reacquired fatigue status data; determine the current sensory stimulation intervention effect based on the changes in fatigue level before and after intervention; adjust subsequent sensory stimulation intervention blocks based on the current sensory stimulation intervention effect.

[0014] Furthermore, the fatigue state data includes eyelid feature data, eye feature data, facial feature data, and behavioral feature data; the eyelid feature data includes eyelid opening and closing degree, duration of eye closure, blinking frequency, and eyelid movement change characteristics; the eye feature data includes eye movement characteristics, fixation behavior characteristics, and gaze deviation characteristics; the facial feature data includes yawning behavior characteristics, mouth movement characteristics, and head posture characteristics; and the behavioral feature data includes steering wheel operation behavior characteristics, pedal control behavior characteristics, and vehicle control behavior characteristics.

[0015] Furthermore, in step one, the fatigue level is determined using one of the following two methods:

[0016] (1) Construct a fatigue evaluation index system based on fatigue state data; conduct a comprehensive evaluation of the fatigue evaluation index system through a preset fuzzy comprehensive evaluation model to determine the fatigue score F of the vehicle driver; determine the fatigue level of the vehicle driver based on the fatigue score.

[0017] When 0 ≤ F < 20, it corresponds to Level 1 fatigue; when 20 ≤ F < 40, it corresponds to Level 2 fatigue; when 40 ≤ F < 60, it corresponds to Level 3 fatigue; when 60 ≤ F < 80, it corresponds to Level 4 fatigue; when 80 ≤ F ≤ 100, it corresponds to Level 5 fatigue.

[0018] (2) Based on the proportion of abnormal eyelid opening and closing, the cumulative duration of continuous eye closure, the degree of deviation of blinking frequency and the degree of fluctuation of eyelid opening and closing, determine the comprehensive risk value of eye movement fatigue of vehicle drivers, compare the comprehensive risk value of eye movement fatigue with the preset fatigue level threshold, and determine the fatigue level of vehicle drivers.

[0019] When 0≤ When <0.20, it corresponds to Level 1 fatigue; when 0.20≤ When <0.40, it corresponds to level 2 fatigue; when 0.40≤ When <0.60, it corresponds to level 3 fatigue; when 0.60≤ When <0.80, it corresponds to level 4 fatigue; when 0.80≤ When the value is ≤1.00, it corresponds to level 5 fatigue.

[0020] Furthermore, the formula for calculating the comprehensive risk value of eye movement fatigue is as follows:

[0021] ;

[0022] in, This indicates the overall risk value of eye movement fatigue within the current testing period; This indicates the percentage of cases with abnormal eyelid opening and closing. T represents the cumulative duration of continuous eye closure; B represents the total duration of the current detection period; and B represents the blinking frequency within the current detection period. This represents the average frequency of blinking for vehicle drivers under normal conditions. This represents the standard deviation of the fluctuation in eyelid opening and closing. The average value of eyelid opening and closing is represented by α, β, γ, and δ, which are weighting coefficients for the proportion of abnormal eyelid opening and closing, the cumulative duration of continuous eye closure, the degree of deviation in blink frequency, and the degree of fluctuation in eyelid opening and closing, respectively, and α+β+γ+δ=1.

[0023] Furthermore, the comprehensive risk value of eye strain will be... The fatigue level of the vehicle driver is determined by comparing the driver's fatigue with a preset fatigue level threshold; the fatigue level determination relationship is expressed as follows:

[0024] ;

[0025] in, This indicates the driver's fatigue level at the current moment; This indicates the fatigue level thresholds set sequentially from low to high; indicator function. The value is 1 when the condition inside the parentheses is true, and 0 when the condition inside the parentheses is false.

[0026] Furthermore, the sensory stimulation intervention groups include groups A, B, C, D, and E; group A includes visual and auditory stimulation, group B includes visual and olfactory stimulation, group C includes visual, auditory, and olfactory stimulation, group D includes visual, auditory, and tactile stimulation, and group E includes visual, auditory, olfactory, tactile, and gustatory stimulation; when the driver is in a level 1 fatigue state, the system calls group A to execute visual and auditory stimulation; when the driver is in a level 2 fatigue state, ... The system invokes area A to provide visual and auditory stimulation. When the driver is in a level 3 fatigue state, the system invokes either area A or area B to provide visual and auditory stimulation, or visual and olfactory stimulation. When the driver is in a level 4 fatigue state, the system invokes either area C or area D to provide visual, auditory, and olfactory stimulation, or visual, auditory, and tactile stimulation. When the driver is in a level 5 fatigue state, the system invokes area E to provide a combined reinforcement intervention using visual, auditory, olfactory, tactile, and gustatory stimulation.

[0027] Furthermore, the sensory stimulus intensity S of the sensory stimulus execution module i Normalized to 0 to 1, i is the sensory stimulation module number, where 0 indicates no stimulus output and 1 indicates the maximum safe stimulus intensity allowed by the current sensory stimulation execution module; the number of sensory stimulation execution modules, sensory stimulation intensity, or sensory stimulation duration are increased progressively from low to high according to the fatigue level; the sensory stimulation intensity corresponding to level 1 fatigue is lower than that corresponding to level 5 fatigue, and the multi-dimensional sensory stimulation strategy corresponding to level 5 fatigue is that the multi-sensory stimulation execution modules are executed synchronously or the multi-sensory stimulation execution modules are executed alternately in a loop;

[0028] The overall fatigue intervention intensity is calculated based on the sensory stimulus intensity and its weight in each sensory stimulus execution module:

[0029] ;

[0030] in, This represents the overall fatigue intervention intensity output to the vehicle driver at the current moment; S i w represents the intensity of the sensory stimulus in the i-th sensory stimulus execution module; i Γ( represents the weight coefficient corresponding to the i-th sensory stimulus execution module;) ) represents a multi-sensory synergistic enhancement function, used to characterize the synergistic fatigue intervention effect produced when multiple sensory stimulus execution modules work together.

[0031] Furthermore, in step three, within the preset feedback collection time after the end of a single round of fatigue intervention, at least one of eyelid feature data, eye feature data, facial feature data, and behavioral feature data is collected; the fatigue level of the vehicle driver is re-determined using a preset fuzzy comprehensive evaluation model or eye movement fatigue comprehensive risk value.

[0032] If the fatigue level after intervention is lower than the fatigue level before intervention, the system can reduce the intensity of sensory stimulation or stop the intervention.

[0033] If the fatigue level after intervention is equal to the fatigue level before intervention, the system can continue to execute the next round of fatigue intervention according to the original sensory stimulation intervention group;

[0034] If the fatigue level after intervention is higher than the fatigue level before intervention, the system can switch to a higher level of sensory stimulation intervention group and shorten the feedback reassessment cycle.

[0035] Furthermore, the update formula for the comprehensive risk value of eye movement fatigue is as follows:

[0036] ;

[0037] in, This indicates the comprehensive risk value of eye movement fatigue for vehicle drivers within the current testing period; This indicates the comprehensive risk value of eye movement fatigue for vehicle drivers in the next testing cycle after fatigue intervention. This represents the sensory stimulus intensity vector of each sensory stimulus execution module at time t; This represents the weight vector corresponding to each sensory stimulus execution module.

[0038] The second objective of this invention is to provide a system for fatigue grading and for fatigue intervention by invoking multiple senses.

[0039] This invention provides a system for fatigue grading and multi-dimensional sensory intervention, which mainly includes the following modules:

[0040] The fatigue detection module is used to acquire fatigue status data of the vehicle driver;

[0041] The fatigue grading module is used to determine the fatigue level of a vehicle driver based on the driver's fatigue status data.

[0042] The sensory stimulation strategy generation module is used to determine the corresponding sensory stimulation intervention group based on the fatigue level of the vehicle driver, and to generate the corresponding sensory stimulation strategy based on the sensory stimulation intervention group.

[0043] The sensory stimulation control module is used to control the corresponding sensory stimulation execution module in the vehicle to execute the sensory stimulation strategy.

[0044] The sensory stimulation execution module is used to execute sensory stimulation strategies according to the commands of the sensory stimulation control module.

[0045] The feedback assessment module is used to acquire fatigue status data after sensory stimulation intervention; reacquire fatigue status data of vehicle drivers; redetermine the fatigue level of vehicle drivers; determine the current sensory stimulation intervention effect based on the changes in fatigue level before and after intervention; and adjust subsequent sensory stimulation intervention blocks based on the current sensory stimulation intervention effect.

[0046] The beneficial effects of this invention are:

[0047] This invention constructs a complete technical solution around four aspects: fatigue detection and grading, multi-dimensional sensory mobilization, sensory stimulation, and feedback reassessment after intervention. It can determine fatigue levels from 1 to 5 based on the multi-source fatigue characteristics of vehicle drivers, and call corresponding sensory stimulation intervention groups according to the fatigue level. After sensory stimulation intervention, the effect of sensory stimulation intervention is judged by reassessing the fatigue level of vehicle drivers and adjusting subsequent sensory stimulation strategies, thereby realizing active identification, graded intervention, and closed-loop optimization of fatigue state.

[0048] This invention enables dynamic adjustment of vehicle driver fatigue status, which not only reduces the influence of subjective judgment on fatigue rating, but also provides clear implementation guidelines for fatigue intervention at different fatigue levels in terms of sensory stimulus intensity, duration, and combination of sensory stimuli. Attached Figure Description

[0049] Figure 1 This is a flowchart of fatigue detection grading based on multi-source features and fuzzy comprehensive evaluation.

[0050] Figure 2 This is a flowchart of multi-dimensional sensory engagement based on fatigue levels.

[0051] Figure 3 This is a flowchart of the feedback reassessment process based on a decrease in fatigue level. Detailed Implementation

[0052] The present invention will be further described in detail below with reference to the accompanying drawings.

[0053] In a first aspect, the present invention provides a method for fatigue grading and using multi-dimensional senses for fatigue intervention.

[0054] This invention provides a method for fatigue grading and multi-dimensional sensory intervention, which mainly includes the following steps:

[0055] Step 1: Fatigue detection grading based on multi-source features and fuzzy comprehensive evaluation; such as Figure 1 As shown, the specific implementation process is as follows:

[0056] Step S101: Obtain fatigue status data of the vehicle driver;

[0057] Once the vehicle is in motion, it continuously collects facial and eye images of the driver, as well as driving behavior data. Facial and eye images can be acquired by onboard cameras, while driving behavior data can be obtained from steering wheel sensors, pedal sensors, vehicle status information acquisition units, and the cockpit perception system.

[0058] Within a detection cycle, fatigue state data such as eyelid feature data, eye feature data, facial feature data, and behavioral feature data are extracted from the raw data. Eyelid feature data mainly includes eyelid opening and closing degree, duration of eye closure, blinking frequency, and eyelid movement change characteristics. Eye feature data mainly includes eye movement characteristics, fixation behavior characteristics, and gaze deviation characteristics. Facial feature data mainly includes yawning behavior characteristics, mouth movement characteristics, and head posture characteristics. Behavioral feature data mainly includes steering wheel operation behavior characteristics, pedal control behavior characteristics, and vehicle control behavior characteristics.

[0059] In some embodiments, the detection period T is set to 30s, and the sliding step size is set to 5s. The sampling frame rate of the vehicle-mounted camera is set to 25fps to 60fps, and the image resolution is not less than 640×480 (pixels); the status refresh period of the cockpit perception system is set to 0.5s to 2s, and the head deflection angle range is 0° to 60°; the refresh period of the vehicle status information acquisition unit is set to 0.1s to 1s, the vehicle speed is 0 to 160km / h, and the continuous driving time is 0 to 480min. The above acquisition parameters ensure that eyelid features, eye features, facial features, and behavioral features can be time-aligned within the same detection period.

[0060] Step S102: Determine the fatigue level of the vehicle driver based on the fatigue data of the vehicle driver.

[0061] The first method: a method based on a pre-defined fuzzy comprehensive evaluation model;

[0062] A fatigue evaluation index system is constructed based on fatigue state data; the fatigue evaluation index system is comprehensively evaluated by a pre-set fuzzy comprehensive evaluation model to determine the fatigue score of the vehicle driver; and the fatigue level of the vehicle driver is determined based on the fatigue score.

[0063] To ensure the reproducibility of the fatigue level determination process, this invention employs a fuzzy comprehensive evaluation method to fuse multi-source fatigue state data. Specifically, the fatigue evaluation factor set can be represented as follows: ,in For eyelid feature data, For eye feature data, For facial feature data, For behavioral characteristic data; the fatigue level evaluation set can be represented as .

[0064] (1) The system comprehensively evaluates various fatigue state data based on pre-defined weight relationships and a fuzzy comprehensive evaluation model. Specifically, the fatigue evaluation factor set... The membership degrees of each factor in the set can be calculated using triangular or trapezoidal membership functions. (Based on the fatigue evaluation factor set...) The first in One factor Corresponding normalized risk value For example, when the first The evaluation range for fatigue level is: Time (r is the fatigue level, It is the interval parameter of the membership function, such as This is the evaluation range for Level 1 fatigue. This is the evaluation range for Level 2 fatigue. This corresponds to the evaluation range for Level 3 fatigue. This corresponds to the evaluation range for level 4 fatigue. That is, the evaluation range for level 5 fatigue), then the... One factor Corresponding normalized risk value membership degree It can be represented as:

[0065] ;

[0066] The fuzzy comprehensive evaluation matrix is ​​composed of the membership degrees of each fatigue evaluation factor and its corresponding normalized risk value. Fuzzy comprehensive evaluation matrix It is a matrix with several rows and columns, where each row represents a fatigue evaluation factor, and each column represents the membership degree of the normalized risk value corresponding to each fatigue evaluation factor (0 represents belonging to that fatigue level, and 1 represents not belonging to that fatigue level). The comprehensive evaluation vector is... ,in This is the set of weight vectors for each fatigue evaluation factor. The fatigue level is determined by the comprehensive evaluation vector. The fatigue level corresponding to the highest membership degree is selected; when there are ties in the highest membership degree, the higher fatigue level is selected to improve the safety redundancy of fatigue warning.

[0067] In some embodiments, the fuzzy comprehensive evaluation matrix The number of rows is 4 and the number of columns is 5, as detailed below:

[0068] ;

[0069] The weight vector can be set as follows: , , , , Let represent the weights of the four fatigue evaluation factors: eyelid feature data, eye feature data, facial feature data, and behavioral feature data, respectively. The corresponding comprehensive evaluation vectors are 1×0.3+1×0.2=0.5, 1×0.35=0.35, 1×0.35+1×0.3+1×0.2+1×0.15=1, 1×0.3=0.3, and 1×0.35+1×0.2=0.55, respectively. We then take the comprehensive evaluation vector... The level corresponding to the highest membership degree is used as the fatigue level, which is level 4 fatigue.

[0070] (2) Based on the results of the procedural scoring described above, fatigue scores can also be calculated. For example, the fatigue score range can be set from 0 to 100 points, and the fatigue levels can be set as Level 1 fatigue, Level 2 fatigue, Level 3 fatigue, Level 4 fatigue, and Level 5 fatigue. The correspondence between fatigue scores and fatigue levels is shown in Table 1.

[0071] Table 1. Correspondence between fatigue score and fatigue level

[0072]

[0073] In some embodiments, within a continuous 60-second detection cycle, the system detected that the driver's average blinking frequency was 28 times per minute, the average duration of eye closure was 0.35 seconds, the number of yawns was 2, and the steering wheel correction frequency was 42 times per minute. The system input these features into a fuzzy comprehensive evaluation model, and after calculation, a fatigue score F of 34 points was obtained. Since 34 points falls within the range of 20 to 40 points, the driver is determined to be in a level 2 fatigue state.

[0074] In some embodiments, while the vehicle is continuously driving on a highway, the system detects that within the last 5 minutes, the PERCLOS index reaches 38%, the average duration of eye closure reaches 0.65 seconds, the number of yawns reaches 5, and the steering wheel correction frequency reaches 68 times per minute. The system inputs the above characteristics into a fuzzy comprehensive evaluation model, and calculates a fatigue score F of 72 points. Since 72 points falls within the range of 60 to 80 points, the system determines that the driver is in a level 4 fatigue state.

[0075] The second method: a method based on the comprehensive risk value of eye movement fatigue;

[0076] Based on data such as the proportion of abnormal eyelid opening and closing, cumulative duration of continuous eye closure, deviation of blinking frequency, and fluctuation of eyelid opening and closing in fatigue state data, the comprehensive risk value of eye movement fatigue for vehicle drivers is determined. The comprehensive risk value of eye movement fatigue is then compared with the preset fatigue level threshold to determine the corresponding fatigue level of the vehicle driver.

[0077] In this invention, the specific formula for calculating the comprehensive risk value of eye movement fatigue is as follows:

[0078] ;

[0079] in, This indicates the overall risk value of eye movement fatigue within the current testing period; This indicates the percentage of cases with abnormal eyelid opening and closing. T represents the cumulative duration of continuous eye closure (in seconds); B represents the total duration of the current detection cycle (in seconds); and B represents the blinking frequency within the current detection cycle (in times / min). This represents the average frequency of blinking by a vehicle driver under normal conditions (unit: blinks / min). This represents the standard deviation of the fluctuation in eyelid opening and closing. α represents the average eyelid opening and closing degree; α, β, γ, and δ represent the weighting coefficients of the proportion of abnormal eyelid opening and closing degree, the cumulative duration of continuous eye closure, the degree of deviation in blinking frequency, and the degree of fluctuation in eyelid opening and closing degree, respectively, and α+β+γ+δ=1.

[0080] In some embodiments, α = 0.35, β = 0.30, γ = 0.20, and δ = 0.15. When the vehicle driver uses the system for the first time, the average eyelid opening and closing angle under non-fatigue conditions within the last 5 minutes can be used as the baseline. The initial reference value; when historical data is available, the average stable eyelid opening and closing angle under the same driving scenario within the last 7 days can be used as the initial reference value. Individualized benchmarks.

[0081] In the embodiment using a five-level fatigue rating system, the comprehensive risk value of eye movement fatigue can be... The fatigue level of the vehicle driver is determined by comparing the driver's fatigue with a preset fatigue level threshold. The fatigue level determination relationship can be expressed as follows:

[0082] ;

[0083] in, This indicates the driver's fatigue level at the current moment; This represents the fatigue level thresholds set sequentially from low to high. This serves as the threshold separating Level 1 fatigue from Level 2 fatigue, and so on. This is the threshold separating level 2 fatigue from level 3 fatigue. This is the threshold separating level 3 and level 4 fatigue. (The threshold between level 4 and level 5 fatigue); indicator function. The value is 1 when the condition inside the parentheses is true, and 0 when the condition inside the parentheses is false.

[0084] Table 2. Correspondence between comprehensive risk value of eye movement fatigue and fatigue level

[0085]

[0086] This invention integrates eyelid features, eye features, facial features, and behavioral features to avoid misjudgments caused by relying solely on a single eye-closing threshold or a single blinking frequency for fatigue detection and grading. By calculating and outputting a fatigue score from 0 to 100 through a fuzzy comprehensive evaluation model, and mapping the fatigue score to fatigue levels 1 to 5, it achieves a quantitative, graded, and continuous expression of fatigue states, providing clear input for subsequent differentiated sensory stimulation interventions.

[0087] Step 2: Multi-dimensional sensory engagement based on fatigue levels; such as... Figure 2As shown, the specific implementation process is as follows:

[0088] Step S201: Determine the corresponding sensory stimulation intervention group based on the fatigue level of the vehicle driver;

[0089] First, determine the sensory stimulation pattern. The sensory stimulation pattern mainly includes: visual stimulation, auditory stimulation, olfactory stimulation, tactile stimulation, and gustatory stimulation.

[0090] Visual stimulation can be achieved through changes in interior lighting strips, ambient lighting, central control display prompts, instrument panel display prompts, head-up display (HUD), or additional lighting devices. Auditory stimulation can be achieved through music playback, voice prompts, or fatigue warning sounds. Olfactory stimulation can be achieved by releasing minty, lemony, coffee, or other stimulating aromatherapy scents. Tactile stimulation can be achieved through air conditioning airflow, temperature adjustment, steering wheel vibration, or seat vibration. Gustatory stimulation can be achieved through in-vehicle stimulating beverages, coffee drinks, energy drinks, mint-based foods, or other stimulating foods.

[0091] Secondly, to avoid the vehicle driver from adapting to sensory stimulation due to long-term use of a single sensory stimulation mode, the present invention combines the above-mentioned sensory stimulation modes to form sensory stimulation intervention blocks.

[0092] In some embodiments, the combination of sensory stimulation intervention blocks can be progressively increased with increasing fatigue level. The combination of each sensory stimulation mode can be represented as follows:

[0093] ;

[0094] Among them, C L The sensory stimulation intervention combination corresponding to the L-level fatigue state is represented by c1 to c5, which correspond to visual stimulation, auditory stimulation, olfactory stimulation, tactile stimulation, and gustatory stimulation, respectively.

[0095] Then, a sensory stimulation intervention group library was constructed based on the combination of various sensory stimulation patterns. The sensory stimulation intervention groups mainly include group A, group B, group C, group D, and group E. Group A mainly includes visual and auditory stimuli; group B mainly includes visual and olfactory stimuli; group C mainly includes visual, auditory, and olfactory stimuli; group D mainly includes visual, auditory, and tactile stimuli; and group E mainly includes visual, auditory, olfactory, tactile, and gustatory stimuli.

[0096] Finally, based on the driver's fatigue level, a corresponding sensory stimulus intervention group can be matched from a pre-defined sensory stimulus intervention group library. Specifically, there is a one-to-one correspondence between fatigue levels and sensory stimulus intervention groups: Level 1 corresponds to Group A, Level 2 to Group B, Level 3 to Group C, Level 4 to Group D, and Level 5 to Group E. This correspondence ensures that fatigue grading results can be directly converted into executable sensory stimulus strategies, avoiding the need for manual re-judgment during the sensory stimulus strategy generation process.

[0097] In some embodiments, the sensory stimulation intervention block selection function can be expressed as:

[0098] G(L) = A L L∈ {1, 2, 3, 4, 5};

[0099] Where G(L) represents the sensory stimulation intervention group selected at time t based on fatigue level, L represents the fatigue level of the vehicle driver at the current time, and A L This represents a sensory stimulation intervention group from group A to group E corresponding to the current fatigue level. For ease of implementation, A1, A2, A3, A4, and A5 can be defined as group A, group B, group C, group D, and group E, respectively.

[0100] Step S202: Generate corresponding sensory stimulation strategies based on the sensory stimulation intervention blocks;

[0101] To directly generate sensory stimulation strategies, the sensory stimulation strategy for each fatigue level can be written into a sensory stimulation strategy table. The sensory stimulation strategy table includes at least the fatigue level number, sensory stimulation strategy number, sensory stimulation execution module number, sensory stimulation execution module weight, sensory stimulation intensity, duration, interval, disabling conditions, and feedback reassessment cycle. Upon receiving a fatigue level, the system retrieves the sensory stimulation strategy from the sensory stimulation strategy table by fatigue level number and sends the retrieved sensory stimulation strategy to the sensory stimulation control module. The sensory stimulation control module then controls the corresponding sensory stimulation execution module to execute the sensory stimulation strategy.

[0102] Specifically, the sensory stimulus intensity S of each sensory stimulus execution module i Normalized to 0 to 1, i is the sensory stimulation module number, where 0 indicates no stimulus output and 1 indicates the maximum safe stimulus intensity allowed by the current sensory stimulation execution module. The number of sensory stimulation execution modules, sensory stimulation intensity, or sensory stimulation duration are increased progressively from low to high according to the fatigue level; the sensory stimulation intensity corresponding to level 1 fatigue is lower than that corresponding to level 5 fatigue, and the multi-dimensional sensory stimulation strategy corresponding to level 5 fatigue is either simultaneous execution of multiple sensory stimulation execution modules or alternating cyclic execution of multiple sensory stimulation execution modules.

[0103] Among them, the intensity of comprehensive fatigue intervention The intensity and weight of the sensory stimuli are determined by the sensory stimuli of each sensory stimulus execution module.

[0104] ;

[0105] in, This represents the overall fatigue intervention intensity output to the vehicle driver at the current moment; n is the number of sensory stimulus execution modules; S i w represents the intensity of the sensory stimulus in the i-th sensory stimulus execution module; i Γ( represents the weight coefficient corresponding to the i-th sensory stimulus execution module;) ) represents a multi-sensory synergistic enhancement function, used to characterize the synergistic fatigue intervention effect produced when multiple sensory stimulus execution modules work together.

[0106] In some embodiments, , The value ranges from 0.05 to 0.20. This refers to the input variables for the multisensory synergistic enhancement term. When visual, auditory, and olfactory stimuli are activated simultaneously, When the intensity of sensory stimulation When the value is between 0.30 and 0.60, the multi-sensory synergistic enhancement term It is used to compensate for the degree of decline in the effect of fatigue intervention after continuous application of a single sensory stimulus.

[0107] Step S203: Control the corresponding sensory stimulus execution module inside the vehicle to execute the sensory stimulus strategy;

[0108] When the driver is in a level 1 fatigue state, the system activates area A to provide visual and auditory stimulation; when the driver is in a level 2 fatigue state, the system activates area A to provide visual and auditory stimulation; when the driver is in a level 3 fatigue state, the system activates either area A or area B to provide visual and auditory stimulation, or visual and olfactory stimulation; when the driver is in a level 4 fatigue state, the system activates either area C or area D to provide visual, auditory, and olfactory stimulation, or visual, auditory, and tactile stimulation; when the driver is in a level 5 fatigue state, the system activates area E to provide a combined reinforcement intervention using visual, auditory, olfactory, tactile, and gustatory stimulation. Details are shown in Table 3.

[0109] Table 3

[0110]

[0111] In some embodiments, when the fatigue level L=3, the sensory stimulation intervention group is determined to be group C, and the sensory stimulation combination is {visual, auditory, olfactory}; the intensity of visual stimulation is set to 0.40, the intensity of auditory stimulation is set to 0.30, and the intensity of olfactory stimulation is set to 0.25; visual and auditory stimulation are started simultaneously, and olfactory stimulation is started after 2 seconds, with a single intervention duration of 15 seconds. After the fatigue intervention ends, the subsequent feedback reassessment process begins within 30 seconds.

[0112] In some embodiments, the system detects that the driver's fatigue score F is 28 points, determining that the driver is in a level 2 fatigue state. The system calls group A to control the interior light strip to switch from a low brightness mode to a high brightness mode, or controls the display screen to display fatigue reminder information; at the same time, it controls the vehicle speakers to play preset voice reminders or refreshing music, which continues for a preset time before entering the subsequent feedback re-evaluation process.

[0113] In some embodiments, the system detects that the driver's fatigue score F is 56 points, determining that the driver is in a level 3 fatigue state. The system can invoke zone C to control the ambient lighting to switch to high brightness mode, play a reminder audio, and simultaneously activate the fragrance release device to release a mint or coffee fragrance, thus stimulating the driver through visual, auditory, and olfactory channels in a coordinated manner.

[0114] In some embodiments, the system detects that the driver's fatigue score F is 88 points, determining that the driver is in a level 5 fatigue state. The system invokes group E to synchronously or in a preset order execute light stimulation, music or voice prompts, aroma release, air conditioning airflow / temperature stimulation, and prompts for stimulating drinks or foods in the vehicle, providing a combined and enhanced intervention for the driver through five sensory stimulation channels.

[0115] Before outputting the sensory stimulation strategy, a safety constraint judgment must be performed. Specifically, if the vehicle is traveling at high speed and the speed is greater than a preset speed threshold (e.g., 100 km / h), then the visual stimulation is restricted to high-frequency flashing, meaning the visual stimulation will not use high-frequency flashing, and the sound pressure level of the auditory stimulation will not exceed a preset sound pressure level threshold (e.g., 70 dB). If the cabin temperature is lower than a first temperature threshold (e.g., 18°C) or higher than a second temperature threshold (e.g., 30°C), then the temperature change range of the tactile stimulation is restricted, meaning the temperature change range of the tactile stimulation is limited to within 2°C. If the vehicle driver sets olfactory or gustatory stimulation to be disabled, then the weight of the corresponding sensory stimulation execution module is automatically set to 0, and the weights of the remaining sensory stimulation execution modules are normalized proportionally.

[0116] This invention constructs different sensory stimuli into groups A to E, and can hierarchically schedule the sensory stimulus execution module, sensory stimulus intensity, and sensory stimulus duration according to the fatigue level. Compared with a single alarm method, this invention can provide differentiated and sustainable fatigue intervention methods under different fatigue levels, and reduce the risk of sensory stimulus adaptation by switching groups, thereby improving the fatigue intervention effect.

[0117] Step 3: Feedback mechanism after fatigue intervention (reassessment based on fatigue level decline); such as... Figure 3 As shown, the specific implementation process is as follows:

[0118] Step S301: Obtain fatigue status data after sensory stimulation intervention;

[0119] After performing multi-dimensional sensory processing based on fatigue levels, the driver's fatigue status data is reacquired. The reacquired fatigue status data includes at least one of eyelid feature data, eye feature data, facial feature data, and behavioral feature data.

[0120] Specifically, eyelid feature data is used to determine whether the driver's eyelid opening and closing has returned to normal; eye feature data is used to determine whether the driver's blinking behavior has become stable; facial feature data is used to determine whether the driver's facial fatigue characteristics have decreased; and behavioral feature data is used to comprehensively assess fatigue changes after sensory stimulation intervention.

[0121] Step S302: Based on the newly acquired fatigue status data, redetermine the fatigue level corresponding to the vehicle driver;

[0122] The first method can be carried out according to the first method in step S102: construct a fatigue evaluation index system based on fatigue state data; conduct a comprehensive evaluation of the fatigue evaluation index system through a preset fuzzy comprehensive evaluation model to determine the fatigue score of the vehicle driver; determine the fatigue level corresponding to the vehicle driver based on the fatigue score.

[0123] The second method: The system determines the comprehensive eye movement fatigue risk value of the driver based on the newly acquired fatigue status data of the driver, compares the comprehensive eye movement fatigue risk value with the preset fatigue level threshold, and determines the new fatigue level corresponding to the driver.

[0124] Specifically, a driver's alertness can naturally decrease with continuous driving and can be improved by external sensory stimulation. The dynamic relationship of a driver's alertness can be expressed as:

[0125] ;

[0126] Where A(t) represents the driver's alertness at time t; λ represents the fatigue decay coefficient, which can range from 0.005 to 0.030 (min). -1 ); Indicates external sensory stimulation intervention input; Φ( This indicates the arousal effect of external sensory stimulation on the driver's alertness.

[0127] After a certain period of fatigue intervention, the comprehensive eye movement fatigue risk value for the vehicle driver in the next detection cycle can be updated based on the cumulative effect of each sensory stimulus execution module during the fatigue intervention process. The update relationship of the comprehensive eye movement fatigue risk value can be expressed as:

[0128] ;

[0129] in, This indicates the comprehensive risk value of eye movement fatigue for vehicle drivers within the current testing period; This indicates the comprehensive risk value of eye movement fatigue for vehicle drivers in the next testing cycle after fatigue intervention. This represents the sensory stimulus intensity vector of each sensory stimulus execution module at time t; This represents the weight vector corresponding to each sensory stimulus execution module; This represents the function that reduces the overall risk of eye strain. Specifically, , The value ranges from 0.10 to 0.35. If the cumulative effect of the intervention... When the level is low, the overall risk value of eye movement fatigue decreases in an approximately linear manner; several pre-cumulative effects... If the intensity continues to increase, the reduction in the overall risk value of eye movement fatigue will gradually approach saturation, in order to avoid the system from over-correcting due to high-intensity stimulation.

[0130] The system can recalculate the comprehensive risk value of eye movement fatigue for vehicle drivers within the current detection period after the feedback collection is completed. Comprehensive evaluation vector and the comprehensive risk value of eye movement fatigue of vehicle drivers in the next testing cycle after fatigue intervention. It will also output a new fatigue level.

[0131] Step S303: Determine the current sensory stimulation intervention effect based on the changes in fatigue level before and after the intervention;

[0132] If the post-intervention fatigue level (new fatigue level) is lower than the pre-intervention fatigue level, the current sensory stimulation intervention is deemed effective; if several post-intervention fatigue levels (new fatigue levels) are the same as or higher than the pre-intervention fatigue level, the current sensory stimulation intervention is deemed insufficient, and the sensory stimulation intervention block adjustment process is initiated.

[0133] Step S304: Adjust the subsequent sensory stimulation intervention blocks based on the current sensory stimulation intervention effect;

[0134] If the fatigue level after intervention (the new fatigue level) is lower than the fatigue level before intervention, the system can reduce the intensity of sensory stimulation or stop the intervention.

[0135] If the fatigue level after intervention (the new fatigue level) is equal to the fatigue level before intervention, the system can continue to execute the next round of fatigue intervention according to the original sensory stimulation intervention group;

[0136] If the fatigue level after intervention (the new fatigue level) is higher than the fatigue level before intervention, the system can switch to a higher-level sensory stimulation intervention block and shorten the feedback reassessment cycle.

[0137] In some embodiments, the feedback reassessment period is set to 30 seconds. After intervention in Zone D, when a vehicle driver's fatigue state has progressed from Level 4 to Level 4, the comprehensive risk value of eye movement fatigue for the vehicle driver within the current detection period is... The value decreased from 0.68 to 0.49, and the comprehensive evaluation vector... When the maximum membership level changes from level 4 to level 3, the fatigue level is updated to level 3, and the sensory stimulation intervention group is downgraded from group D to group C; the comprehensive risk value of eye movement fatigue for vehicle drivers during the current testing period. When the score is less than 0.20 for two consecutive feedback reassessment cycles, the system stops the multisensory stimulation execution module intervention and only retains the low-intensity visual cue for group A.

[0138] The system can record the execution information and recovery results of each sensory stimulation intervention. The execution information mainly includes the fatigue level before intervention, the fatigue level after intervention, the sensory stimulation intervention group used, the duration of intervention, and the degree of fatigue reduction. Based on these records, the system can establish an individualized intervention profile for vehicle drivers and prioritize the use of sensory stimulation intervention groups with better historical recovery results for subsequent fatigue levels of the same or similar levels.

[0139] In some embodiments, the driver's fatigue score F before the intervention was 84 points, corresponding to a level 5 fatigue state. After the system performed the intervention in Block E, a re-evaluation was conducted, resulting in a post-intervention fatigue score F of 67 points, corresponding to a level 4 fatigue state. Since the post-intervention fatigue level was lower than the pre-intervention fatigue level, the sensory stimulation intervention was deemed effective. The system may reduce the intensity of the sensory stimulation or maintain a low intensity of sensory stimulation monitoring based on the driving environment.

[0140] In some embodiments, the driver is in a level 3 fatigue state before intervention. The system initially calls group A for visual and auditory stimulation. After a preset time, the system reassesses the driver's fatigue level. If the driver is still in a level 3 fatigue state, the sensory stimulation intervention in group A is deemed insufficient, and the system switches to group B or other candidate groups to introduce different combinations of sensory stimulation to continue the intervention.

[0141] This invention directly determines the effectiveness of sensory stimulation intervention by re-collecting fatigue data and re-evaluating the driver's fatigue level after intervention. By using fatigue level as the success criterion, the evaluation results are clear and interpretable. By switching sensory stimulation intervention blocks and adjusting sensory stimulation strategies when intervention is ineffective, the system can avoid remaining on ineffective sensory stimulation strategies after a single failed intervention, thus forming a closed-loop control process of "fatigue detection - fatigue grading - sensory stimulation intervention - feedback re-evaluation - strategy adjustment".

[0142] Secondly, the present invention provides a system for fatigue grading and using multi-dimensional senses for fatigue intervention, which is mainly applied to the method for fatigue grading and using multi-dimensional senses for fatigue intervention provided in the first aspect.

[0143] The present invention provides a fatigue grading system that utilizes multi-dimensional senses for fatigue intervention, which mainly includes: a fatigue detection module, a fatigue grading module, a sensory stimulation strategy generation module, a sensory stimulation control module, a sensory stimulation execution module, and a feedback evaluation module.

[0144] The fatigue detection module is mainly used to acquire fatigue status data of vehicle drivers;

[0145] The fatigue grading module is mainly used to determine the fatigue level of a vehicle driver based on the driver's fatigue status data.

[0146] The sensory stimulation strategy generation module is mainly used to determine the corresponding sensory stimulation intervention group based on the fatigue level of the vehicle driver, and to generate the corresponding sensory stimulation strategy based on the sensory stimulation intervention group.

[0147] The sensory stimulation control module is mainly used to control the corresponding sensory stimulation execution module in the vehicle to execute the sensory stimulation strategy.

[0148] The sensory stimulus execution module is mainly used to execute sensory stimulus strategies according to the commands of the sensory stimulus control module.

[0149] The feedback assessment module is mainly used to acquire fatigue status data after sensory stimulation intervention; reacquire fatigue status data of vehicle drivers; redetermine the fatigue level of vehicle drivers; determine the current sensory stimulation intervention effect based on the changes in fatigue level before and after intervention; and adjust subsequent sensory stimulation intervention blocks based on the current sensory stimulation intervention effect.

[0150] In some embodiments, the sensory stimulation control module may be integrated into the vehicle cockpit controller or the vehicle driver monitoring controller, or it may be set up independently.

[0151] In some embodiments, the sensory stimulation execution module mainly includes a visual stimulation device, an auditory stimulation device, an olfactory stimulation device, a tactile stimulation device, and a gustatory stimulation device. The visual stimulation device mainly includes at least one of the following: interior light strip, ambient lighting, central control display screen, instrument display screen, HUD head-up display system, and additional lighting devices; the auditory stimulation device mainly includes at least one of the following: in-vehicle speaker, cabin audio system, headrest speaker, or directional audio device; the olfactory stimulation device mainly includes at least one of the following: fragrance generator, volatile fragrance device, odor release module, or air conditioning linkage release module; the tactile stimulation device mainly includes at least one of the following: air conditioning airflow stimulation device, temperature control device, steering wheel vibration device, seat vibration device, or seatbelt vibration device; the gustatory stimulation device mainly includes at least one of the following: in-vehicle beverage reminder device, stimulating beverage storage device, stimulating food reminder device, or automatic dispensing device.

[0152] This invention systematically integrates fatigue detection, fatigue grading, sensory stimulation strategy generation, sensory stimulation execution, and feedback evaluation to form a unified fatigue intervention and control architecture. It can reuse existing vehicle hardware resources such as lights, audio, air conditioning, displays, and seats, reducing deployment costs, while expanding fatigue intervention capabilities by adding fragrance release, beverage or stimulating food prompts, etc.

[0153] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for fatigue grading and utilizing multi-dimensional senses for fatigue intervention, characterized in that, Includes the following steps: Step 1: Fatigue detection grading based on multi-source features and fuzzy comprehensive evaluation; Acquire driver fatigue data; determine driver fatigue level based on driver fatigue data; Step 2: Multi-sensory mobilization based on fatigue level; The corresponding sensory stimulation intervention group is determined based on the fatigue level of the vehicle driver; the corresponding sensory stimulation strategy is generated based on the sensory stimulation intervention group; and the corresponding sensory stimulation execution module in the vehicle is controlled to execute the sensory stimulation strategy. The sensory stimulation intensity S of the sensory stimulation execution module i Normalized to 0 to 1, i is the sensory stimulation module number, where 0 indicates no stimulus output and 1 indicates the maximum safe stimulus intensity allowed by the current sensory stimulation execution module; the number of sensory stimulation execution modules, sensory stimulation intensity, or sensory stimulation duration are increased progressively from low to high according to the fatigue level; the sensory stimulation intensity corresponding to level 1 fatigue is lower than that corresponding to level 5 fatigue, and the multi-dimensional sensory stimulation strategy corresponding to level 5 fatigue is that the multi-sensory stimulation execution modules are executed synchronously or the multi-sensory stimulation execution modules are executed alternately in a loop; The overall fatigue intervention intensity is calculated based on the sensory stimulus intensity and its weight in each sensory stimulus execution module: ； in, This represents the overall fatigue intervention intensity output to the vehicle driver at the current moment; S i w represents the intensity of the sensory stimulus in the i-th sensory stimulus execution module; i Γ( represents the weight coefficient corresponding to the i-th sensory stimulus execution module;) ) represents a multi-sensory synergistic enhancement function, used to characterize the synergistic fatigue intervention effect produced when multiple sensory stimulus execution modules work together; Step 3: Reassess based on feedback of decreased fatigue level; Acquire fatigue status data after sensory stimulation intervention; redetermine the fatigue level of the vehicle driver based on the reacquired fatigue status data; determine the current sensory stimulation intervention effect based on the changes in fatigue level before and after intervention; adjust subsequent sensory stimulation intervention blocks based on the current sensory stimulation intervention effect.

2. The method for fatigue grading and multi-dimensional sensory intervention according to claim 1, characterized in that, The fatigue state data includes eyelid feature data, eye feature data, facial feature data, and behavioral feature data; the eyelid feature data includes eyelid opening and closing degree, duration of eye closure, blinking frequency, and eyelid movement change characteristics; the eye feature data includes eye movement characteristics, fixation behavior characteristics, and gaze deviation characteristics; the facial feature data includes yawning behavior characteristics, mouth movement characteristics, and head posture characteristics; and the behavioral feature data includes steering wheel operation behavior characteristics, pedal control behavior characteristics, and vehicle control behavior characteristics.

3. The method for fatigue grading and multi-dimensional sensory intervention according to claim 1, characterized in that, In step one, the fatigue level is determined using one of the following two methods: (1) Construct a fatigue evaluation index system based on fatigue state data; conduct a comprehensive evaluation of the fatigue evaluation index system through a preset fuzzy comprehensive evaluation model to determine the fatigue score F of the vehicle driver; The fatigue level of the vehicle driver is determined based on the fatigue score. When 0 ≤ F < 20, it corresponds to Level 1 fatigue; when 20 ≤ F < 40, it corresponds to Level 2 fatigue; when 40 ≤ F < 60, it corresponds to Level 3 fatigue; when 60 ≤ F < 80, it corresponds to Level 4 fatigue; when 80 ≤ F ≤ 100, it corresponds to Level 5 fatigue. (2) Based on the proportion of abnormal eyelid opening and closing, the cumulative duration of continuous eye closure, the degree of deviation of blinking frequency and the degree of fluctuation of eyelid opening and closing, determine the comprehensive risk value of eye movement fatigue of vehicle drivers, compare the comprehensive risk value of eye movement fatigue with the preset fatigue level threshold, and determine the fatigue level of vehicle drivers. When 0≤ When <0.20, it corresponds to Level 1 fatigue; when 0.20≤ When <0.40, it corresponds to level 2 fatigue; when 0.40≤ When <0.60, it corresponds to level 3 fatigue; when 0.60≤ When <0.80, it corresponds to level 4 fatigue; when 0.80≤ When the value is ≤1.00, it corresponds to level 5 fatigue. This indicates the overall risk value of eye movement fatigue within the current testing period.

4. The method for fatigue grading and multi-dimensional sensory intervention according to claim 3, characterized in that, The formula for calculating the comprehensive risk value of eye movement fatigue is as follows: ； in, This indicates the overall risk value of eye movement fatigue within the current testing period; This indicates the percentage of cases with abnormal eyelid opening and closing. T represents the cumulative duration of continuous eye closure; B represents the total duration of the current detection period; and B represents the blinking frequency within the current detection period. This represents the average frequency of blinking for a vehicle driver under normal conditions. This represents the standard deviation of the fluctuation in eyelid opening and closing. The average value of eyelid opening and closing is represented by α, β, γ, and δ, which are weighting coefficients for the proportion of abnormal eyelid opening and closing, the cumulative duration of continuous eye closure, the degree of deviation in blink frequency, and the degree of fluctuation in eyelid opening and closing, respectively, and α+β+γ+δ=1.

5. The method for fatigue grading and multi-dimensional sensory intervention according to claim 4, characterized in that, Comprehensive risk value of eye strain The fatigue level of the vehicle driver is determined by comparing the driver's fatigue with a preset fatigue level threshold; the fatigue level determination relationship is expressed as follows: ； in, This indicates the driver's fatigue level at the current moment; This indicates the fatigue level thresholds set sequentially from low to high; indicator function. The value is 1 when the condition inside the parentheses is true, and 0 when the condition inside the parentheses is false.

6. The method for fatigue grading and multi-dimensional sensory intervention according to claim 1, characterized in that, The sensory stimulation intervention groups include groups A, B, C, D, and E. Group A includes visual and auditory stimuli; group B includes visual and olfactory stimuli; group C includes visual, auditory, and olfactory stimuli; group D includes visual, auditory, and tactile stimuli; and group E includes visual, auditory, olfactory, tactile, and gustatory stimuli. When the driver is in a level 1 fatigue state, the system calls group A to execute visual and auditory stimuli; when the driver is in a level 2 fatigue state, the system... The system invokes area A to provide visual and auditory stimulation. When the driver is in a level 3 fatigue state, the system invokes area A or area B to provide visual and auditory stimulation, or visual and olfactory stimulation. When the driver is in a level 4 fatigue state, the system invokes area C or area D to provide visual, auditory, and olfactory stimulation, or visual, auditory, and tactile stimulation. When the driver is in a level 5 fatigue state, the system invokes area E to provide a combined reinforcement intervention involving visual, auditory, olfactory, tactile, and gustatory stimulation.

7. The method for fatigue grading and multi-dimensional sensory intervention according to claim 1, characterized in that, In step three, within the preset feedback collection time after the end of a single round of fatigue intervention, at least one of eyelid feature data, eye feature data, facial feature data, and behavioral feature data is collected; the fatigue level of the vehicle driver is re-determined using a preset fuzzy comprehensive evaluation model or eye movement fatigue comprehensive risk value. If the fatigue level after intervention is lower than the fatigue level before intervention, the system can reduce the intensity of sensory stimulation or stop the intervention. If the fatigue level after intervention is equal to the fatigue level before intervention, the system can continue to execute the next round of fatigue intervention according to the original sensory stimulation intervention group; If the fatigue level after intervention is higher than the fatigue level before intervention, the system can switch to a higher level of sensory stimulation intervention group and shorten the feedback reassessment cycle.

8. The method for fatigue grading and multi-dimensional sensory intervention according to claim 7, characterized in that, The formula for updating the comprehensive risk value of eye movement fatigue is: ； in, This indicates the comprehensive risk value of eye movement fatigue for vehicle drivers during the current testing period; This indicates the comprehensive risk value of eye movement fatigue for vehicle drivers in the next testing cycle after fatigue intervention. This represents the sensory stimulus intensity vector of each sensory stimulus execution module at time t; This represents the weight vector corresponding to each sensory stimulus execution module.

9. A system for fatigue grading and multi-dimensional sensory intervention as described in any one of claims 1-8, characterized in that, include: The fatigue detection module is used to acquire fatigue status data of the vehicle driver; The fatigue grading module is used to determine the fatigue level of a vehicle driver based on the driver's fatigue status data. The sensory stimulation strategy generation module is used to determine the corresponding sensory stimulation intervention group based on the fatigue level of the vehicle driver, and to generate the corresponding sensory stimulation strategy based on the sensory stimulation intervention group. The sensory stimulation control module is used to control the corresponding sensory stimulation execution module in the vehicle to execute the sensory stimulation strategy. The sensory stimulation execution module is used to execute sensory stimulation strategies according to the commands of the sensory stimulation control module. The feedback assessment module is used to acquire fatigue status data after sensory stimulation intervention; reacquire fatigue status data of vehicle drivers; redetermine the fatigue level of vehicle drivers; determine the current sensory stimulation intervention effect based on the changes in fatigue level before and after intervention; and adjust subsequent sensory stimulation intervention blocks based on the current sensory stimulation intervention effect.

Images